Research map / Video / Video segmentation and tracking
Learning To Detect Mirrors From Videos via Dual Correspondences
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add the dual correspondence module, including both stages, rather than using a single-frame convolutional baseline (ablation) | VMD-Net · VMD-D test | IoU: 0.424 -> 0.567 | Helped |
| Add the dual correspondence module, including both stages, rather than using a three-frame convolutional baseline (ablation) | VMD-Net · VMD-D test | IoU: 0.446 -> 0.567 | Helped |
| Add the first stage of the dual correspondence module (ablation) | VMD-Net · VMD-D test | IoU: 0.524 -> 0.567 | Helped |
| Add the second stage of the dual correspondence module (ablation) | VMD-Net · VMD-D test | IoU: 0.533 -> 0.567 | Helped |
| Use different frames instead of three copies of the same frame (ablation) | VMD-Net · VMD-D test | IoU: 0.525 -> 0.567 | Helped |
| Use both randomly selected and adjacent frames rather than only adjacent frames (ablation) | VMD-Net · VMD-D test | IoU: 0.539 -> 0.567 | Helped |
| Use intra-frame and inter-frame correspondences for video mirror detection | VMD-Net · VMD-D test | IoU: 0.539 -> 0.567 | Helped |
Closest papers by meaning
- Effective Video Mirror Detection with Inconsistent Motion Cues (CVPR 2024 · similarity 0.93)
- ZOOM: Learning Video Mirror Detection with Extremely-Weak Supervision (AAAI 2024 · similarity 0.91)
- Detect Any Mirrors: Boosting Learning Reliability on Large-Scale Unlabeled Data with an Iterative Data Engine (CVPR 2025 · similarity 0.88)
- Weakly-Supervised Mirror Detection via Scribble Annotations (AAAI 2024 · similarity 0.88)
- Multi-View Dynamic Reflection Prior for Video Glass Surface Detection (AAAI 2024 · similarity 0.85)
- MIRROR: Make Your Object-Level Multi-View Generation More Consistent with Training-Free Rectification (ICML 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.